calm-plumber-49567
10/20/2025, 1:21 AMAccess request denied because roundtrip failed
If the offending pod is killed (or the deployment is restarted) things go back to normal for 4-6 hours but that is a large amount of instability that has arisen in what has otherwise been a very stable deployment over the last 18 months.
We've been seeing these issues running v0.40.6 , but I'm upgrading us to v0.40.9
Thanks all for any advice or insight!